MNRF high frequency upgrade decisions

R. Ekers - May 1996

Summary (See memo AT 50/40 - 26 April 1996)

• 12mm proceeds on all antennas as in MNRF proposal
• 3.5mm MMIC 85-95GHz (more tuning range if possible), 6 systems
• Extend 3mm surface from 15 to 22m *
• No SIS receivers *
• No 2.7mm except for MMIC demonstrator project
• 3 new stations *
• LO upgrade *
• VLBI upgrades as in MNRF proposal

Preamble

The AT director will make a recommendation to the Australia Telescope Steering Committee, and if accepted this becomes part of the formal DIST/MNRF agreement which we hope to submit by late May 96. Note that the ATNF operating budget is now very tight and the MNRF funding includes no additional operating funds, so upgrade options which either decrease or have neutral impact on the operating budget are favored. (See * in summary)

High frequency upgrade

12mm

The science case for the 12mm upgrade remains essentially as proposed. All 6 ATCA antennas plus Mopra, Hobart and Ceduna will include a 12mm receiver. One spare receiver will be included. The main remaining scientific issue will be the frequency range. 16-26 GHz has been requested but this is likely to be too large for optimal Tsys and polarization purity. ATUC should consider the scientific importance of lines especially at the extremes of this range and the importance of the push to the lower end of the range for the continuum band.

3 v 7mm

The initial discussions and the original MNRF proposal considered a (85-115GHz) system using the inner 15m of the antennas. The science case was dominated by considerations of the 115GHz transition of CO at the upper end of this band. We built a 85-115GHz SIS receiver for Mopra to evaluate the potential of the site and the antenna, and to regain experience at mm wavelengths. Although the receiver performance was excellent, the site and antenna were disappointing at 115GHz and indicated that the AT would be 30-40 times less sensitive at 115GHz than at lower frequencies. Much of the unexpected degradation in performance was related to the effect of the antenna optics and spillover with the 15m illumination. Also note that the antenna performance is dropping off exponentially at 115GHz. Further difficulties were expected as a result of the EW array configuration which forces us to observe at large hour angle and hence low elevation, and the lack of short spacing information which is exacerbated by only illuminating 15 of the 22m. The credibility of the 115GHz (2.7mm) part of the proposal was also questioned by a skeptical ESO committee containing experienced mm observers.

The case for 7mm (35-50GHz) was re-examined and found to be preferable in many respects and community discussion of these options started (see discussion on the ATNF web site www.atnf.csiro.au/_mail_archive/MNRF/). The most important aspect of the new discussion was the introduction of the case for 3.5mm (80-95GHz) in its own right and we thank John Storey in particular for this emphasis.

The change of emphasis from the 2.7 to the 3.5mm end of the band was important since comparison of the 7mm and 3.5mm science clearly favors 3.5mm. With 15m aperture the 3.5mm system would still be 8-10 times less sensitive than 7mm but this would be offset by the greater line strength (x22) and larger number of transitions. Imaging considerations also favor the 3.5mm system since we would have considerable difficulty using the ATCA to image the more extended and complex CO emission at 2.7mm compared with the 3.5mm lines which generally trace regions of high density which are more compact. Although the 3.5mm (15m aperture) case is clearly better than the case for 2.7mm, the 7 v 3.5mm decision would still be a close call given the advantages of 7mm for VLBI, non-thermal and CMB observations, and the difficulty of imaging at 3.5mm using an EW array of 15m antennas with 30m minimum spacing and 15m increments.

If we illuminate the full 22m aperture at 3.5mm we gain another factor of 2-3 in S/N, we fully sample the UV plane (the 15m increments were designed for 22m aperture illumination), and we go some way to recovering the short spacing information by mosaicing. The cost of the required design study and 3.5mm surface extensions to 22m for the 5 ATCA antennas on the 3km track is an additional $0.9M but this does have the extra advantage of not increasing operating costs. Requirements on pointing and subreflector stability are eased a little at 3.5mm. To extend the 3.5mm surface to 22m within the MNRF funding umbrella we propose the following:

• Use 3.5mm MMIC or HEMT receivers (not SIS) on only the 5 ATCA antennas on the 3km track. These receivers to be included in the 12mm package.
• no 7mm receivers
• drop plans for further mm upgrades of Mopra. This includes beam switches, focal plane arrays and wide band correlator. We will however probably extend the 3mm Mopra surface to 22m to provide a test bed for the ATCA modifications. Note that some of the Mopra mm developments may still be pursued through the international funding component and collaboration with ESO/SEST.
• postpone purchase of hydrogen maser time standards for Hobart and Parkes.

New ATCA stations

The case for 3 new stations is strong for all high frequency upgrades. We are considering a variation in which one station has 6 (rather than 4) piers allowing shortest spacing of either 26 or 34m. This will increase the range of declination for which shadowing is minimized, and will considerably improve the short spacing information.

LO upgrade for ATCA

As proposed in the original MNRF proposal.

Water vapor radiometers

If results of interferometer tests with the first 2 systems are successful we will construct 3 more to equip all ATCA antennas with radiometers. Final decision to be made in 1997.

VLBI

All funding for Ceduna and upgrading current systems to 12mm are included as proposed. The loss of the 7mm system which covers the SiO maser will have negative impact, and it is unclear whether 3.5mm will ever become a useful VLBI frequency. The most valuable option foreseen is for at least 2-3 7mm systems on the ATCA (but not at Mopra). This is not currently within the MNRF funding profile. If we lose the hydrogen masers at Hobart or Parkes (both on loan) we will have a serious problem. This could possibly be covered through our normal future facilities funding, or by purchasing additional less expensive Chinese masers.

Options still open and issues for discussion? (as of May 1996)

• Reshape ATCA antennas for off axis performance (could be done as part of the upgrade to 22m)
• 3mm receiver and surface extension for ATCA 6km antenna
• 3mm surface extended from 15 to 22m at Mopra
• Actual frequency range at 3.5mm. 85-100GHz is probably too great - which end should we clip?
• Drop dual polarization to achieve more frequency agility at 3.5mm
• 3 additional water vapor radiometers
• 7mm at Mopra
• 7mm interferometer (2-3 elements) on ATCA
• H-masers to replace existing foreign owned masers at Parkes and Hobart